Physicist review: Radiant_One_Youtube (Step 2)

Reviewer: PhD physicist role. Sources: the schematic PNG (viewed directly) [read: page image p.1], facts.json, claims.md, the digest, and the Investigator's sites.md and research/ notes (no files in research/files/; there is no deep/ folder because the project is one image).

Source gap (repeated from claims.md). The folder holds one image, a schematic screenshot from the YouTube channel radiant_one. No video, transcript, measurement or author identity is in the folder. Everything about measurements (30 mA, temperatures, 4 mW per LED) comes from the Investigator's reading of the author's YouTube pages and auto-generated transcripts [web, via research/2026-10-04-youtube.md; I did not re-watch the videos]. Calculations below are mine from the schematic and those numbers. Where I rely on [memory] I say so; no verdict depends on a medium/low-confidence [memory] fact, so no web checks were needed beyond the Investigator's.

Corrections to Step 1

1. facts.json says each LED string is "shunted by an anti-parallel 1N4148". The image [read: page image p.1] shows two 1N4148 diodes, one drawn along the top and one along the bottom, each spanning the whole two-string LED group between the L2 node and the earth-rod node, in opposite directions. So it is one anti-parallel diode pair across the combined array, not one diode per string. Orientation of the LEDs themselves is not drawn.
2. facts.json treats the orange vertical bar as "apparently ... coupling between L1 and L2". The Investigator's finding that the PNG is a frame from video GrdNrgEx_2 [web] makes it more likely a video-editing overlay. Treat the bar as having no electrical meaning.
3. Claim 3 in claims.md says the equal-current condition is "presented as the tuned operating condition". The Investigator's transcript note shows more: it is a calibration the author sets by moving the clip, not a measured outcome. Equal current is the input to the experiment, not a finding of it. I carry this into the verdicts below.
4. facts.json calls the claim "extracting net energy ... conflicts with conservation of energy". The author's own descriptions say "No overunity performance is suggested or implied" in most videos [web, research notes]. Both statements (Moray "ground energy" framing and the no-overunity disclaimer) are the author's; Step 1 gave only the first. Both are kept here.
5. The Investigator's "30 mA x 11-13 V" input is the Investigator's arithmetic and fine (0.33-0.39 W), but the supply is set at 11-13 V on the schematic; the 12 V battery in GndNrgEx_9 [web] gives about 0.36 W at 30 mA. I use 12 V.

What the circuit is (mainstream reading)

Plain-language summary for the reader. An NPN transistor (MPSA06) with a feedback network (C1 = 470 pF in series with Lb = 1.4 uH) and a collector coil L1 (8.6 uH) forms an RF oscillator of the "Slayer exciter" / Stiffler "SEC-15" family; the author says so himself [web, GufTw6ECoz4 transcript]. A second coil L2 (26.7 uH) in the collector path feeds a load: two strings of 26 white LEDs in parallel, with an anti-parallel 1N4148 pair across them, whose far end goes to an earth rod.

Key numbers I computed: - Resonant frequency f = 1/(2 pi sqrt(LC)) of the Lb-C1 branch: 1/(2 pi sqrt(1.4e-6 x 470e-12)) = 6.2 MHz. With L1 (8.6 uH) and 470 pF: 2.5 MHz. The actual oscillation frequency was not measured in anything I have; it is some MHz value. This is a rough estimate, not a measurement. - Forward drop of a white LED is about 2.9 V [memory, high], so a 26-LED string needs about 75 V across it before it conducts. From a 12 V supply this needs voltage step-up. - Stray capacitance needed to resonate L2 at 6.2 MHz: C = 1/((2 pi f)^2 L) = 1/((3.9e7)^2 x 26.7e-6) = about 25 pF. That is a typical size for the capacitance between a bench circuit, a person, a table and the room's surroundings and earth [memory, medium; order of magnitude only]. At 6 MHz, 25 pF has reactance about 1 kOhm.

Mainstream explanation of how the LEDs light: L2 and the stray capacitance to earth form a series resonant circuit at about the driving frequency. In a series resonant circuit the voltage across L2 and across the capacitor can be many times the drive voltage (Q times). The current closes through the earth rod, the ground, and the stray (displacement) capacitance back to the circuit, so the earth rod is the lead that completes the loop; displacement current is real current for circuit purposes (Maxwell's addition to Ampere's law; Griffiths ch. 7). The LEDs rectify, so over a cycle the net current must be zero (a capacitor passes no DC); the anti-parallel 1N4148 pair carries the half-cycle the LEDs block, at 0.7 V rather than 75 V. This also explains the author's observation that "ground energy appears to stop when it encounters a 1N4148" [web, GndNrgEx_10 description]: a single diode in series with a capacitive path just rectifies, leaving no way for the net current to close.

Energy source in this reading: the 12 V supply. The earth rod provides a path, a "counterpoise" in radio language. Stiffler's own SEC documents describe the earth as "a counterpoise" (Google snippet only, not read) [web, research/2026-10-04-energeticforum.md]: that is a mainstream radio-engineering description of the same wire that radiant_one calls a source of "ground energy". Both are the authors' characterisations.

Energy-balance calculation using the author's own numbers

Author's reported numbers [web]: supply current 30 mA earthed and unearthed; 12 V battery; "each LED is only drawing 4 mW"; 52 LEDs. - Input power: P_in = 12 V x 0.030 A = 0.36 W. - Reported LED power: 52 x 4 mW = 0.208 W (the "4 mW per LED" is read from the video description; how it was measured is not documented, and RF makes meters unreliable, as the author warns). - Ratio: 0.208 / 0.36 = 58 percent. Not an excess: output is below input. At face value the numbers show an RF driver with 58 percent of its input reaching the LEDs, a plausible figure for a class-C style oscillator and resonant load. Nothing here requires an energy source other than the battery. - The author's heat test [web, GndNrgEx_3]: transistor +1.2 C above ambient when unearthed and +0.7 C when earthed after 5 minutes. A smaller temperature rise when 208 mW is going to LEDs means less power is dissipated in the transistor when it is loaded: mainstream expectation (an unloaded oscillator tank re-circulates power and radiates or dissipates it in the transistor and coils). The author reads this as "most of the energy is consumed generating RF" when unearthed. This is consistent with the mainstream reading, and the author does not claim it as excess energy. - Why equal current is no surprise: the clip is moved until unearthed current equals earthed current. That selects a coupling at which the oscillator's loading by the (open) earth branch equals its loading by the LED array. Equal supply current then means equal input power in both states; the same 0.36 W goes to RF/losses when unearthed and to LEDs plus losses when earthed. Conservation of energy is satisfied in both states. The "unchanged current" is therefore a result of the tuning procedure, not evidence of a free source.

What the author did not measure (from the Investigator's notes): RF power, LED light output (GndNrgEx_4 promised to measure it but is not on the public list), ground current, supply ripple or true input power using a time-averaged V x I with an RF-safe method. Note that a DC ammeter on a supply lead can be wrong when RF is present (the author's own warning), and a battery's DC current is usually read properly only with a decoupling capacitor across the supply.

Claim 1. A "Ground Energy Extraction" circuit can draw usable electrical energy from the earth through an earth rod.

a. Mathematics internally consistent? No equations are given by the author; the schematic is self-consistent as a circuit. The series-resonant picture above fits it with no extra source. The statement that "a small energy input from the driver produces a much larger response from the ground electrons" [web, GndNrgEx_10] is not quantified; "much larger" cannot be checked against any number in the record. Not testable as a mathematical statement.

b. Consistent with established experiments? The earth is not a known source of usable power at the levels needed. Measured natural earth currents (telluric currents) and earth-ionosphere atmospheric fields are real, but of microamperes per square metre and volts per metre of order 100 V/m near ground in fair weather [memory, high]. A small rod collecting from that supplies nanowatts to microwatts, many orders below 208 mW. Another established effect: an earthed antenna or counterpoise lets an RF circuit couple energy through the ground; that is what Marconi-type transmitters and Tesla-type exciters rely on [memory, high]. The reported LED power (about 0.2 W) is below the 0.36 W input, so even at face value the data are consistent with the supply as the source.

c. Tested? By the author alone, informally (current meter, an IR thermometer, a temperature probe). No neutral test is on record. The Investigator found no independent replication, no forum discussion located (overunity.com thread of 2015 not found), and one YouTube comment saying "You didn't show the circuit working" (2018) [web]. Other YouTube videos of "radiant energy grounding" (Okki Moeljadi; George Chaniotakis) are leads only, not read. Absence of mainstream testing is the normal record for a hobby video; it is not evidence either way.

d. If true, what would change? If a net output exceeded supply input in a controlled measurement, energy conservation would be violated unless a new ambient source of order 100 mW per square metre of rod were identified, which would also be seen in simple earth-potential measurements. The weak form (the earth rod completes a return path) changes nothing.

Verdict (split). - Strong form (net usable energy from the ground, more than the supply puts in): Never properly tested (no neutral, controlled power measurement of input, output and ground contribution on record; not dismissed by anyone as far as the record shows; the author's own disclaimer says no overunity is suggested). On the author's own numbers the output is 58 percent of input, so the data do not support the strong form either, and the physics gives no known source of the size needed. I am not calling it "Refuted by calculation": the numbers needed for a calculated refutation (true RF input and output powers) are not in the record. - Weak form (the earth rod provides a return path, lets the LEDs light, and the driver is not seen to draw more current): Established as ordinary RF circuit behaviour (series resonance with stray capacitance, plus rectification).

Claim 2. An MPSA06 RF oscillator (L1 8.6 uH, Lb 1.4 uH, C1 470 pF, 1 Meg, 11-13 V) coupled by L2 (26.7 uH) can drive an earth-rod-connected load of two 26-LED strings with anti-parallel 1N4148.

a. Mathematics consistent? As a circuit, yes. Slayer-exciter-type oscillators with one MPSA06 on 12 V are described in the Energetic Forum builds of 2011-12 (12 V, 30-75 mA, lighting LEDs and fluorescent tubes) [web, research/2026-10-04-energeticforum.md]. The step-up from 12 V to the 75 V needed to light 26 LEDs is possible by resonance as computed (a Q of about 10 would suffice). The values (8.6, 1.4, 26.7 uH; 470 pF) are consistent with MHz operation.

b. Consistent with experiments? Yes: forum builds show the same family lighting LEDs and tubes at tens of mA. Lighting an LED array from an RF field with a long wire or an earth connection is old and well known (e.g. fluorescent tubes lit near a Tesla coil) [memory, high].

c. Tested? Only by the author and forum builders of the base circuit (SEC-15 / Slayer). I found nothing independent on the earthed LED array itself.

d. If true, what changes? Nothing. It is ordinary RF engineering.

Verdict: Plausible, untested as a specific build (the circuit is consistent with known physics and its base design is widely built; no independent build of this exact LED array is documented). The step-up mechanism is Established in principle.

Claim 3. Supply current with the earth rod connected can be tuned (alligator clip near L2) to equal the current without it.

a. Mathematics consistent? Yes, and trivially so: the clip is a capacitive probe between the base clamp diodes' centre tap and L2. Moving it changes the feedback coupling from L2 into the base and hence the oscillator drive and supply current. A continuous adjustment can always make two currents equal (intermediate value theorem). Consistent.

b. Consistent with experiments? Yes. Any oscillator's supply current depends on loading and feedback.

c. Tested? Only by the author [web: 30 mA both states; Investigator's transcript note]. The observation that equal-current can be set is easily reproduced by anyone who builds it. It does not test any energy claim.

d. If true, what changes? Nothing; equal input power in two states does not show that the output is free. The meaningful comparison is output power at the LEDs against input power, which the record gives as 58 percent. The author's temperature test (transistor cooler when earthed) is consistent with energy conservation.

Verdict: Established as a reproducible tuning result, with the important qualifier that it has no bearing on net energy. It is a tuned condition, not an independent discovery. As evidence for Claim 1 it is Not testable as stated (equal input says nothing about whether a further source supplies the LEDs).

Claim 4. The circuit produces enough RF that meters and oscilloscopes can be damaged or read incorrectly.

a. Consistent? Yes; this is the author's warning on the schematic [read: page image p.1]. b. Experiments? Well established: strong RF fields cause rectification in meter input stages and induce voltage in probe leads; infrared thermometer reading "-50" [web, GndNrgEx_3 transcript] is a typical RF-upset symptom. The author recounts that the RF "can destroy a multimeter or oscilloscope, I know that from personal experience" [web]. c. Tested by the author's own experience; the author's workaround (switch off and measure at once) is sensible but gives only a lower bound on heating. d. Nothing changes.

Verdict: Established. Consequence for Claim 1: the author's own warning means all his meter readings (including "4 mW per LED" and the supply current) have unquantified RF error. That cuts both ways: the figures are not strong evidence for an excess and not for the contrary.

Investigator's findings, weighed (kept in the record whatever the verdict)

• Inventor's own work [web]: twelve videos 2015-2018; eight on ground energy; tuning, heat and LED power statements as above; the author's framing in Moray's 1920s-30s "radiant energy" device and Tesla single-wire transmission. The Moray device itself has a documented demonstration history and claimants [memory, high that it exists; medium on details]; I do not rely on it here. GndNrgEx_4, 6, 7 and 8 are missing from the channel: the Investigator's competing explanations are that they were never made, deleted, made private or renumbered; the Wayback Machine could decide. The gap matters because _4 was promised to measure light output, the very data an energy balance needs.
• Single-source evidence: the author's temperatures (23.7 vs 23.2 C against 22.5 C ambient) and "4 mW per LED" are one source (the author's videos), uncorroborated and unreplicated, with no stated uncertainty. They are quoted above and used in the balance; at face value they do not show an excess. A 0.5 C difference between two readings from an IR thermometer after RF exposure is small for any firm conclusion (the author mentions the instrument was disturbed by RF) [web].
• Claim in GndNrgEx_9 that the fibreglass table is "partly responsible for the flow of magnetic-dielectric interaction": not quantified. Mainstream view: a dielectric table changes stray capacitance slightly; it would matter for the resonance computed above. Untested in the record.
• "One Wire Energy" videos (10 m of 30 gauge wire, receiver brighter when earthed; metal enclosure around the exciter) [web]: a single wire with a ground at the receiver carries a displacement-current loop closed via stray capacitances; a "full metal jacket" shields the near field but the wire still radiates and couples. This is consistent with ordinary RF theory and is not, by itself, evidence of longitudinal waves. Not tested with a proper return measurement.
• Related work (Stiffler SEC exciters, Slayer exciter, Energetic Forum builds, YouTube radiant-grounding demos, Tesla/Moray precedents) [web]: the circuit lineage is documented and mainstream-compatible; Stiffler's own description of the earth as counterpoise [web, snippet only; documents not read] supports the mainstream reading.
• Historical/record gap: author anonymous; no patent, paper or company found (Investigator). Not a gap that suggests suppression: a hobbyist channel with 469 subscribers is the expected record.

What would settle it (for the reader to try or specify)

1. Measure input power by recording supply voltage and current with a shunt resistor and oscilloscope (or by averaging with a large decoupling capacitor at the supply), both earthed and unearthed.
2. Measure LED power directly: probe voltage across and current through the string with an RF-safe current probe or small shunt, or measure light output with a calibrated lux or photodiode against a DC-driven reference LED.
3. Run the whole thing from a battery with the earth rod disconnected from the room's mains earth, and replace the earth rod with an equal-capacitance floating metal plate or a wire to a distant capacitor. If the LEDs light as brightly with a floating plate of the same area, the "ground" is a capacitor plate, not an energy source.
4. Check whether the output exceeds input: if P_LED / P_in < 1 with honest uncertainty, the circuit is an inefficient RF converter.

Verdict summary